You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We use eksctl to create managed Nodegroups for our cluster.
What happened?
We noticed the following error when checking cloudtrail for the CreateAutoScalingGroup API call (08:13:09 (UTC+00:00)):
"You must use a valid fully-formed launch template. Value (eks-88c9a977-44e8-7220-1bc8-dbbca45af78b) for parameter iamInstanceProfile.name is invalid. Invalid IAM Instance Profile name"
The API call for the createInstanceProfile was successful at 08:13:06 (UTC+00:00).
The CreateAutoScalingGroup API call will retry after few seconds and it will work successfully at 08:13:40 (UTC+00:00)
However, in some of our clusters, the CreateAutoScalingGroup API call did not retry and ASG was not created.
This happens randomly in some of the clusters and its hard to replicate.
Logs
There are no specific logs to mention, the only thing we could see from cloudtrail is that the createAutoscalingGroup API call was not retried and the ASG was not created for some of the clusters.
The solution was that we deleted the Nodegroup cloudformation and re-created it again, and this time it worked.
Anything else we need to know?
We did reach out to AWS support and they mentioned that at the AMI side there is a propagation delay and that the two API calls had only few seconds difference.
They suggested to add a delay between these API calls.
Please let us know how this can be done or if this can be fixed?
Versions
$ eksctl info
v0.183.0
The text was updated successfully, but these errors were encountered:
Hello inesshz 👋 Thank you for opening an issue in eksctl project. The team will review the issue and aim to respond within 1-5 business days. Meanwhile, please read about the Contribution and Code of Conduct guidelines here. You can find out more information about eksctl on our website
What were you trying to accomplish?
We use eksctl to create managed Nodegroups for our cluster.
What happened?
We noticed the following error when checking cloudtrail for the CreateAutoScalingGroup API call (08:13:09 (UTC+00:00)):
"You must use a valid fully-formed launch template. Value (eks-88c9a977-44e8-7220-1bc8-dbbca45af78b) for parameter iamInstanceProfile.name is invalid. Invalid IAM Instance Profile name"
The API call for the createInstanceProfile was successful at 08:13:06 (UTC+00:00).
The CreateAutoScalingGroup API call will retry after few seconds and it will work successfully at 08:13:40 (UTC+00:00)
However, in some of our clusters, the CreateAutoScalingGroup API call did not retry and ASG was not created.
This happens randomly in some of the clusters and its hard to replicate.
Logs
There are no specific logs to mention, the only thing we could see from cloudtrail is that the createAutoscalingGroup API call was not retried and the ASG was not created for some of the clusters.
The solution was that we deleted the Nodegroup cloudformation and re-created it again, and this time it worked.
Anything else we need to know?
We did reach out to AWS support and they mentioned that at the AMI side there is a propagation delay and that the two API calls had only few seconds difference.
They suggested to add a delay between these API calls.
Please let us know how this can be done or if this can be fixed?
Versions
v0.183.0
The text was updated successfully, but these errors were encountered: